Timing of intervention with antipsychotic medication may influence long-term outcome in schizophrenia in a manner that is poorly understood. This study evaluated psychopathology, its factor structure, and cognitive dysfunction in older patients with chronic schizophrenia in relation to the intervals from onset of psychosis to initiation of treatment with antipsychotics, and from initiation of antipsychotic treatment to current assessments. The subjects were 129 patients with schizophrenia, many of whom became ill in the preneuroleptic era. Their current psychopathology was assessed using the Positive and Negative Syndrome Scale, and its factor structure examined using principal component analysis. Current general and executive cognitive function was evaluated using the Mini-Mental State Examination and the Executive Interview, respectively. Using multiple regression modelling, increasing duration of initially unmedicated psychosis, but not the much longer duration of subsequently treated illness, was the primary predictor of psychomotor poverty (negative symptoms) but not of reality distortion or disorganisation over the three domains of psychopathology resolved; duration of initially unmedicated psychosis marginally predicted the severity of general, but not of executive, cognitive dysfunction. Delayed intervention with antipsychotics appears associated with poorer long-term course in terms of increased severity of psychopathology in the psychomotor poverty domain.